Portmanteau tests for periodic ARMA models with dependent errors

Abstract

In this article, we derive the asymptotic distributions of residual and normalized residual empirical autocovariances and autocorrelations of (parsimonious) periodic autoregressive moving-average (PARMA) models under the assumption that the errors are uncorrelated but not necessarily independent. We then deduce the modified portmanteau statistics. We establish the asymptotic behavior of the proposed statistics. It is shown that the asymptotic distribution of the modified portmanteau tests is that of a weighted sum of independent chi-squared random variables, which can be different from the usual chi-squared approximation used under independent and identically distributed assumption on the noise. We also propose another test based on a self-normalization approach to check the adequacy of PARMA models. A set of Monte Carlo experiments and an application to financial data are presented.